Data for the Probability Flow Inference (PFI) package
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This repository contains datasets necessary for the reprodution of examples notebooks in the package pfi (https://github.com/vchz/pfi).This data corresponds to single-cell RNA sequencing, over a few days, of HSC as they differentiate in-vitro into various blood cells. The two first datasets are in human, and the last one in mice. The data contained in this repository was previous published. If you use this data, cite the original papers (listed below). https://www.science.org/doi/10.1126/science.adi8577This data can be accessed at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE305370.We used the file 'GSE305370_rna_combined_with_velocity_and_refined_annotations.h5ad' which contains CITE-seq data preprocessed with a log-transform and a normalization. We reverse engineered this transformation with the following code: X = (np.exp(X) - 1) all_min = np.array([X[i,X[i,:] > 0].min() for i in range(X.shape[0])]) size_fac = 1/all_min X = size_fac[:,None]*X print(np.max(X - np.round(X))) X = X.astype(np.float32) With the print(np.max(X - np.round(X))) we check with that with this inverse transformation we recover integer values everywhere. https://www.nature.com/articles/s41467-021-27159-xThis data can be accessed at https://zenodo.org/records/5291737. We used the AnnData object 'SCANPY_anndata_object.h5ad', for which we replaced the preprocessed data by the raw count data which was stored in the layer 'matrix'. https://www.science.org/doi/10.1126/science.aaw3381This data can be accessed at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE140802 and https://github.com/AllonKleinLab/paper-data/blob/master/Lineage_tracing_on_transcriptional_landscapes_links_state_to_fate_during_differentiation/.We reconstructed the clonal information from the information available on the github repository of the AllonKleinLab. For a given cell, the number of fates observed in its lineage are number in the obs column 'num_fates' (most lineages only show one fate). The most represented fate of a lineage is in the obs column 'max_fate'. Finally, the proportion of fates observed in a lineage for a given cell are stored in obs columns 'p_[name_of_fate]'. Overall this information reconstructs the state-fate map in Fig. 2.C of the article.
Fields of Study & Subjects
Single-cell and spatial transcriptomics
Gene Regulatory Network Analysis
Topological and Geometric Data Analysis
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 26 March 2026
- Language
- EN
- Document type
- Dataset